Prof. Xudong Jiang
Nanyang Technological University, Singapore

Speech Title: TBA
Abstract: TBA

Bio-Sketch: Xudong Jiang received the B.Eng. and M.Eng. from the University of Electronic Science and Technology of China (UESTC), and the Ph.D. degree from Helmut Schmidt University, Hamburg, Germany. From 1986 to 1993, he was a Lecturer with UESTC, where he received two Science and Technology Awards from the Ministry for Electronic Industry of China. From 1998 to 2004, he was with the Institute for Infocomm Research, A-Star, Singapore, as a Lead Scientist and the Head of the Biometrics Laboratory, where he developed a system that achieved the most efficiency and the second most accuracy at the International Fingerprint Verification Competition in 2000. He joined Nanyang Technological University (NTU), Singapore, as a Faculty Member, in 2004, and served as the Director of the Centre for Information Security from 2005 to 2011. Currently, he is a professor in NTU. Dr Jiang holds 7 patents and has authored over 200 papers with 2 papers in Nature Communications, 20 papers in Pattern Recognition and over 40+ papers in the IEEE journals, including 6 papers in IEEE Transactions on Pattern Analysis and Machine Intelligence and 14 papers in IEEE Transactions on Image Processing. Four of his papers have been listed as the top 1% highly cited papers in the academic field of Engineering by Essential Science Indicators. He served as IFS TC Member of the IEEE Signal Processing Society from 2015 to 2017, Associate Editor for IEEE Signal Processing Letter from 2014 to 2018, Associate Editor for IEEE Transactions on Image Processing from 2016 to 2020 and the founding editorial board member for IET Biometrics form 2012 to 2019. Dr Jiang is currently an IEEE Fellow and serves as Senior Area Editor for IEEE Transactions on Image Processing and Editor-in-Chief for IET Biometrics. His current research interests include image processing, pattern recognition, computer vision, machine learning, and biometrics.

Prof. Tetsuya Shimamura
Saitama University, Japan

Speech Title: Emotion Recognition from Speech, Image and Brain Data
Abstract: Understanding human emotions is a crucial step toward artificial intelligence, and thus this topic is the subject of extensive research. Emotion recognition (ER) task is often reduced to that of pattern recognition. ER is accomplished through some sensing data such as speech, face image and brain signals. The use of speech for ER, which is often referred to as speech emotion recognition (SER), is becoming popular, recently. Based on several datasets, sophisticated algorithms have been developed. In this talk, a powerful technique for the purpose of SER is shown, in which the use of air-conducted (AC) speech as well as bone-conducted (BC) speech is considered. Several convolutional neural network(CNN) based architectures for training and testing are discussed and how data augmentation, dropout and regularizations are useful is demonstrated through experiments. The difference between AC and BC speech is also discussed, and the usefulness of BC speech is unveiled.

Emotion recognition from face image is a very competitive work. A transfer leaning approach is introduced as one of the most powerful emotion recognizers in which only face image data is given as input. A performance comparison among the state-of the art approaches is shown, visualizing that a performance improvement is obtained as the number of layers employed in deep learning architecture increases.

The brain signal is another potent information source to recognize emotion. Electroencephalography (EEG) is a preferred brain signal, where the crucial and challenging task is accurately extracting features from complex EEG signals using appropriate computational intelligence or machine learning techniques. Recent methods mostly use EEG channel connectivity features to identify the emotion. Specifically, it is useful to construct a connectivity feature map (CFM) for ER methods. An enhanced CFM is shown as one proposal, which uses partial mutual information (PMI) by introducing an extra third channel to expose more information and strengthen the feature extraction ability of ER. The proposed technique calculates the PMI-based connectivity features for each pair of EEG channels and presents CFM in 2D and 3D forms. CNN is used to classify emotion. It is demonstrated that the proposed one outperforms the existing related recent methods.

Bio-Sketch: Tetsuya Shimamura received his B.E., M.E. and PhD degrees in Electrical Engineering from Keio University in Japan, in 1986, 1988 and 1991, respectively. In 1991, he joined Saitama University in Japan, where he is currently a Professor of Graduate School of Science and Engineering.

He was Head of Department of Information and Computer Sciences at Saitama University in 2012 and 2013, and Dean of Information Technology Center in 2014 and 2015. In 1995 and 1996, he joined Loughborough University, UK, and The Queen’s University of Belfast, UK, respectively, as a visiting Professor.

His research interests are in digital signal processing and its applications to speech, audio, image and communication systems. A various range of research is covered such as speech analysis, speech enhancement, image quality assessment, image restoration, wireless communication, sensor network and cognitive radio. He has published over 500 refereed journal articles and international conference proceedings papers. He is an author or co-author of eight books, and a member of the organizing committee of several international conferences.

He has received IEEE Pacific Rim Conference on Communications, Computers and Signal Processing, Gold Paper Award, in 2012, WSEAS International Conference on Multimedia Systems and Signal Processing, Best Paper Award, in 2013, IEEE IFOST, Best Paper Award, in 2014, and IEEE ICCE-TW, Best Paper Award, 2025. Also, he is a recipient of Journal of Signal Processing, Best Paper Award, in 2013, 2015, and 2016, and Yahagi Commemorative Award of Journal of Signal Processing, in 2018. He is an IEEE senior member.